From 35779eda7a2ad52f2bbbe2e1251ad8ddb4bac7d1 Mon Sep 17 00:00:00 2001
From: vallovic <vallovic@gmail.com>
Date: Fri, 19 Feb 2021 14:10:10 +0000
Subject: [PATCH 1/9] Dealing with RTP latest changes
---
youtube_dl/extractor/rtp.py | 75 +++++++++++++++++++++++++++----------
1 file changed, 55 insertions(+), 20 deletions(-)
diff --git a/youtube_dl/extractor/rtp.py b/youtube_dl/extractor/rtp.py
index 02986f442..665c86f14 100644
--- a/youtube_dl/extractor/rtp.py
+++ b/youtube_dl/extractor/rtp.py
@@ -6,22 +6,27 @@ from ..utils import (
determine_ext,
js_to_json,
)
+from ..compat import (
+ compat_b64decode,
+ compat_urllib_parse_unquote,
+)
+
+import re
class RTPIE(InfoExtractor):
- _VALID_URL = r'https?://(?:www\.)?rtp\.pt/play/p(?P<program_id>[0-9]+)/(?P<id>[^/?#]+)/?'
+ _VALID_URL = r'https?://(?:www\.)?rtp\.pt/play/(.*\/)?p(?P<program_id>[0-9]+)/(?P<id>[^/?#]+)/?'
_TESTS = [{
- 'url': 'http://www.rtp.pt/play/p405/e174042/paixoes-cruzadas',
- 'md5': 'e736ce0c665e459ddb818546220b4ef8',
+ 'url': 'https://www.rtp.pt/play/p117/e476527/os-contemporaneos',
'info_dict': {
- 'id': 'e174042',
- 'ext': 'mp3',
- 'title': 'Paixões Cruzadas',
- 'description': 'As paixões musicais de António Cartaxo e António Macedo',
+ 'id': 'e476527',
+ 'ext': 'mp4',
+ 'title': 'Os Contemporâneos Episódio 1 - RTP Play - RTP',
+ 'description': 'Os Contemporâneos, um programa de humor com um olhar na sociedade portuguesa!',
'thumbnail': r're:^https?://.*\.jpg',
},
}, {
- 'url': 'http://www.rtp.pt/play/p831/a-quimica-das-coisas',
+ 'url': 'https://www.rtp.pt/play/p510/aleixo-fm',
'only_matching': True,
}]
@@ -29,30 +34,60 @@ class RTPIE(InfoExtractor):
video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id)
- title = self._html_search_meta(
- 'twitter:title', webpage, display_name='title', fatal=True)
+ title = self._html_search_regex(r'<title>(.+?)</title>', webpage, 'title')
+
+ # Get JS object
+ js_object = self._search_regex(r'(?s)RTPPlayer *\( *({.+?}) *\);', webpage, 'player config')
+
+ json_string_for_config = ''
+
+ # Verify JS object since it isn't pure JSON and maybe it needs some decodings
+ for line in js_object.splitlines():
+ stripped_line = line.strip()
+
+ # If JS object key is 'file'
+ if re.match('file ?:', stripped_line):
+ if 'decodeURIComponent' in stripped_line:
+ # 1) The file URL is inside object and with HLS encoded...
+ hls_encoded = re.match(r"[^[]*\[([^]]*)\]", stripped_line).groups()[0]
+ hls_encoded = hls_encoded.replace('"', '').replace('\'', '').replace(',', '')
+ decoded_file_url = compat_b64decode(
+ compat_urllib_parse_unquote(
+ hls_encoded.replace('"', '').replace(',', ''))).decode('utf-8')
+
+ # Insert the decoded HLS file URL into pure JSON string
+ json_string_for_config += '\nfile: "' + decoded_file_url + '",'
+ else:
+ # 2) ... or it's a direct M3U8 file
+ json_string_for_config += '\n' + line
+
+ elif not stripped_line.startswith("//") and not re.match('fileKey ?:', stripped_line):
+ # Ignore commented lines and 'fileKey' entry since it is no longer supported by RTP
+ json_string_for_config += '\n' + line
+
+ # Finally send pure JSON string for JSON parsing
+ config = self._parse_json(json_string_for_config, video_id, js_to_json)
+
+ # config = self._parse_json(self._search_regex(
+ # r'(?s)RTPPlayer ?\( ?({.+?})\);', webpage,
+ # 'player config'), video_id, js_to_json)
- config = self._parse_json(self._search_regex(
- r'(?s)RTPPlayer\(({.+?})\);', webpage,
- 'player config'), video_id, js_to_json)
file_url = config['file']
ext = determine_ext(file_url)
+
if ext == 'm3u8':
- file_key = config.get('fileKey')
+ # Download via m3u8 file
formats = self._extract_m3u8_formats(
file_url, video_id, 'mp4', 'm3u8_native',
- m3u8_id='hls', fatal=file_key)
- if file_key:
- formats.append({
- 'url': 'https://cdn-ondemand.rtp.pt' + file_key,
- 'preference': 1,
- })
+ m3u8_id='hls')
+
self._sort_formats(formats)
else:
formats = [{
'url': file_url,
'ext': ext,
}]
+
if config.get('mediaType') == 'audio':
for f in formats:
f['vcodec'] = 'none'
From a85625977d39b4ad5bb9b399773700fa20db7587 Mon Sep 17 00:00:00 2001
From: vallovic <vallovic@gmail.com>
Date: Sat, 20 Feb 2021 20:42:33 +0000
Subject: [PATCH 2/9] 'Estudo em Casa' wasn't working since RTP has a lot of
ways of dealing with their code
---
youtube_dl/extractor/rtp.py | 11 ++++++-----
1 file changed, 6 insertions(+), 5 deletions(-)
diff --git a/youtube_dl/extractor/rtp.py b/youtube_dl/extractor/rtp.py
index 665c86f14..02716c172 100644
--- a/youtube_dl/extractor/rtp.py
+++ b/youtube_dl/extractor/rtp.py
@@ -58,7 +58,7 @@ class RTPIE(InfoExtractor):
# Insert the decoded HLS file URL into pure JSON string
json_string_for_config += '\nfile: "' + decoded_file_url + '",'
else:
- # 2) ... or it's a direct M3U8 file
+ # 2) ... or the file URL is not encoded so keep it that way
json_string_for_config += '\n' + line
elif not stripped_line.startswith("//") and not re.match('fileKey ?:', stripped_line):
@@ -68,11 +68,12 @@ class RTPIE(InfoExtractor):
# Finally send pure JSON string for JSON parsing
config = self._parse_json(json_string_for_config, video_id, js_to_json)
- # config = self._parse_json(self._search_regex(
- # r'(?s)RTPPlayer ?\( ?({.+?})\);', webpage,
- # 'player config'), video_id, js_to_json)
+ # Check if file URL is directly a string or is still inside object
+ if isinstance(config['file'], str):
+ file_url = config['file']
+ else:
+ file_url = config['file']['hls']
- file_url = config['file']
ext = determine_ext(file_url)
if ext == 'm3u8':
From b955e0a5dc208abaa5f7d64e7228c46bb8a3af63 Mon Sep 17 00:00:00 2001
From: vallovic <vallovic@gmail.com>
Date: Sun, 21 Feb 2021 14:05:42 +0000
Subject: [PATCH 3/9] RTP devs like to try out different approaches
---
youtube_dl/extractor/rtp.py | 13 ++++++++-----
1 file changed, 8 insertions(+), 5 deletions(-)
diff --git a/youtube_dl/extractor/rtp.py b/youtube_dl/extractor/rtp.py
index 02716c172..5ba26268e 100644
--- a/youtube_dl/extractor/rtp.py
+++ b/youtube_dl/extractor/rtp.py
@@ -15,11 +15,11 @@ import re
class RTPIE(InfoExtractor):
- _VALID_URL = r'https?://(?:www\.)?rtp\.pt/play/(.*\/)?p(?P<program_id>[0-9]+)/(?P<id>[^/?#]+)/?'
+ _VALID_URL = r'https?://(?:(?:(?:www\.)?rtp\.pt/play/(?P<subarea>.*/)?p(?P<program_id>[0-9]+)/(?P<episode_id>e[0-9]+/)?)|(?:arquivos\.rtp\.pt/conteudos/))(?P<id>[^/?#]+)/?'
_TESTS = [{
'url': 'https://www.rtp.pt/play/p117/e476527/os-contemporaneos',
'info_dict': {
- 'id': 'e476527',
+ 'id': 'os-contemporaneos',
'ext': 'mp4',
'title': 'Os Contemporâneos Episódio 1 - RTP Play - RTP',
'description': 'Os Contemporâneos, um programa de humor com um olhar na sociedade portuguesa!',
@@ -51,9 +51,12 @@ class RTPIE(InfoExtractor):
# 1) The file URL is inside object and with HLS encoded...
hls_encoded = re.match(r"[^[]*\[([^]]*)\]", stripped_line).groups()[0]
hls_encoded = hls_encoded.replace('"', '').replace('\'', '').replace(',', '')
- decoded_file_url = compat_b64decode(
- compat_urllib_parse_unquote(
- hls_encoded.replace('"', '').replace(',', ''))).decode('utf-8')
+ if 'atob' in stripped_line:
+ decoded_file_url = compat_b64decode(
+ compat_urllib_parse_unquote(
+ hls_encoded.replace('"', '').replace(',', ''))).decode('utf-8')
+ else:
+ decoded_file_url = compat_urllib_parse_unquote(hls_encoded)
# Insert the decoded HLS file URL into pure JSON string
json_string_for_config += '\nfile: "' + decoded_file_url + '",'
From 7f40887b29f97d4729e344b04e76456f02c0c33a Mon Sep 17 00:00:00 2001
From: vallovic <vallovic@gmail.com>
Date: Tue, 13 Apr 2021 21:03:59 +0100
Subject: [PATCH 4/9] Ignore more comments in RTP source code
---
youtube_dl/extractor/rtp.py | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/youtube_dl/extractor/rtp.py b/youtube_dl/extractor/rtp.py
index 5ba26268e..1c487749b 100644
--- a/youtube_dl/extractor/rtp.py
+++ b/youtube_dl/extractor/rtp.py
@@ -64,8 +64,8 @@ class RTPIE(InfoExtractor):
# 2) ... or the file URL is not encoded so keep it that way
json_string_for_config += '\n' + line
- elif not stripped_line.startswith("//") and not re.match('fileKey ?:', stripped_line):
- # Ignore commented lines and 'fileKey' entry since it is no longer supported by RTP
+ elif not stripped_line.startswith("//") and not re.match('fileKey ?:', stripped_line) and not re.match('.*extraSettings ?:', stripped_line):
+ # Ignore commented lines, 'fileKey' entry since it is no longer supported by RTP and also 'extraSettings'
json_string_for_config += '\n' + line
# Finally send pure JSON string for JSON parsing
From ff47d11269d6cffb36c8b86cf8f5e8effa881670 Mon Sep 17 00:00:00 2001
From: vallovic <vallovic@gmail.com>
Date: Wed, 21 Apr 2021 16:24:26 +0100
Subject: [PATCH 5/9] Ignore EVEN more comments in RTP source code
---
youtube_dl/extractor/rtp.py | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/youtube_dl/extractor/rtp.py b/youtube_dl/extractor/rtp.py
index 1c487749b..2f9360e12 100644
--- a/youtube_dl/extractor/rtp.py
+++ b/youtube_dl/extractor/rtp.py
@@ -34,6 +34,10 @@ class RTPIE(InfoExtractor):
video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id)
+
+ # Remove JS multi-line comments from webpage source
+ webpage = re.sub(r'(\/\*.*\*\/)', '', webpage, flags=re.DOTALL)
+
title = self._html_search_regex(r'<title>(.+?)</title>', webpage, 'title')
# Get JS object
From d30180d4f3e1c0f1f01f7a3aa3d3fa250ce23b15 Mon Sep 17 00:00:00 2001
From: vallovic <vallovic@gmail.com>
Date: Fri, 23 Apr 2021 23:02:04 +0100
Subject: [PATCH 6/9] Consider multi-part videos in filename output
---
youtube_dl/extractor/rtp.py | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/youtube_dl/extractor/rtp.py b/youtube_dl/extractor/rtp.py
index 2f9360e12..da30a7789 100644
--- a/youtube_dl/extractor/rtp.py
+++ b/youtube_dl/extractor/rtp.py
@@ -40,6 +40,14 @@ class RTPIE(InfoExtractor):
title = self._html_search_regex(r'<title>(.+?)</title>', webpage, 'title')
+ # Replace irrelevant text in title
+ title = title.replace(' - RTP Play - RTP', '')
+
+ # Check if it's a video split in parts, if so add part number to title
+ part = self._html_search_regex(r'section\-parts.*<span.*>(.+?)</span>.*</ul>', webpage, 'part', default=None)
+ if part:
+ title = f'{title} {part}'
+
# Get JS object
js_object = self._search_regex(r'(?s)RTPPlayer *\( *({.+?}) *\);', webpage, 'player config')
From 5a7243b741f2f9a746b0955de8bae2f558efd074 Mon Sep 17 00:00:00 2001
From: vallovic <vallovic@gmail.com>
Date: Thu, 6 May 2021 23:20:20 +0100
Subject: [PATCH 7/9] Working around new URLs
---
youtube_dl/extractor/rtp.py | 3 +++
1 file changed, 3 insertions(+)
diff --git a/youtube_dl/extractor/rtp.py b/youtube_dl/extractor/rtp.py
index da30a7789..d421f2f1a 100644
--- a/youtube_dl/extractor/rtp.py
+++ b/youtube_dl/extractor/rtp.py
@@ -70,6 +70,9 @@ class RTPIE(InfoExtractor):
else:
decoded_file_url = compat_urllib_parse_unquote(hls_encoded)
+ # Workaround for new behaviour
+ decoded_file_url = decoded_file_url.replace('streaming-vod.rtp.pt/hls/', 'streaming-ondemand.rtp.pt/').replace('.mp4/', '/')
+
# Insert the decoded HLS file URL into pure JSON string
json_string_for_config += '\nfile: "' + decoded_file_url + '",'
else:
From 6bde4a13779bd8b7202240cf5036be4a2ca242b2 Mon Sep 17 00:00:00 2001
From: vallovic <vallovic@gmail.com>
Date: Sat, 28 Aug 2021 21:10:18 +0100
Subject: [PATCH 8/9] Fix 'RTP Arquivos' downloads, allow subtitles download,
more tests, general cleanup
---
youtube_dl/extractor/rtp.py | 135 +++++++++++++++++++++++++++---------
1 file changed, 102 insertions(+), 33 deletions(-)
diff --git a/youtube_dl/extractor/rtp.py b/youtube_dl/extractor/rtp.py
index d421f2f1a..ffa915075 100644
--- a/youtube_dl/extractor/rtp.py
+++ b/youtube_dl/extractor/rtp.py
@@ -3,6 +3,7 @@ from __future__ import unicode_literals
from .common import InfoExtractor
from ..utils import (
+ ExtractorError,
determine_ext,
js_to_json,
)
@@ -17,13 +18,53 @@ import re
class RTPIE(InfoExtractor):
_VALID_URL = r'https?://(?:(?:(?:www\.)?rtp\.pt/play/(?P<subarea>.*/)?p(?P<program_id>[0-9]+)/(?P<episode_id>e[0-9]+/)?)|(?:arquivos\.rtp\.pt/conteudos/))(?P<id>[^/?#]+)/?'
_TESTS = [{
- 'url': 'https://www.rtp.pt/play/p117/e476527/os-contemporaneos',
+ 'url': 'https://www.rtp.pt/play/p117/e563265/os-contemporaneos',
'info_dict': {
'id': 'os-contemporaneos',
'ext': 'mp4',
- 'title': 'Os Contemporâneos Episódio 1 - RTP Play - RTP',
+ 'title': 'Os Contemporâneos Episódio 1',
'description': 'Os Contemporâneos, um programa de humor com um olhar na sociedade portuguesa!',
- 'thumbnail': r're:^https?://.*\.jpg',
+ 'thumbnail': r're:^https?://.*\.(jpg|png)'
+ },
+ }, {
+ 'url': 'https://www.rtp.pt/play/p8157/e541212/telejornal',
+ 'info_dict': {
+ 'id': 'telejornal',
+ 'ext': 'mp4',
+ 'title': 'Telejornal de 01 Mai 2021 PARTE 1',
+ 'description': 'A mais rigorosa seleção de notícias, todos os dias às 20h00. De segunda a domingo, João Adelino Faria e José Rodrigues dos Santos mostram-lhe o que de'
+ },
+ }, {
+ 'url': 'https://www.rtp.pt/play/p6646/e457262/grande-entrevista',
+ 'info_dict': {
+ 'id': 'grande-entrevista',
+ 'ext': 'mp4',
+ 'title': 'Grande Entrevista Episódio 7 - de 19 Fev 2020',
+ 'description': 'Bruno Nogueira - É um dos mais originais humoristas portugueses e de maior êxito! Bruno Nogueira na Grande Entrevista com Vítor Gonçalves.'
+ },
+ }, {
+ 'url': 'https://www.rtp.pt/play/estudoemcasa/p7776/e539826/portugues-1-ano',
+ 'info_dict': {
+ 'id': 'portugues-1-ano',
+ 'ext': 'mp4',
+ 'title': 'Português - 1.º ano , aula 45 - 27 Abr 2021 - Estudo Em Casa - RTP',
+ 'description': 'A História do Pedrito Coelho, de Beatrix Potter. O dígrafo \'lh\' - A História do Pedrito Coelho, de Beatrix Potter. O dígrafo \'lh\'.'
+ },
+ }, {
+ 'url': 'https://www.rtp.pt/play/zigzag/p5449/e385973/banda-zig-zag',
+ 'info_dict': {
+ 'id': 'banda-zig-zag',
+ 'ext': 'mp4',
+ 'title': 'Banda Zig Zag Episódio 1 - Zig Zag Play - RTP',
+ 'description': 'A Amizade é o Nosso Mel - Zig: é a menina que além de tocar também canta. Adora aprender palavras novas e adora ler. Gosta de fazer palavras cruzadas'
+ },
+ }, {
+ 'url': 'https://arquivos.rtp.pt/conteudos/liga-dos-ultimos-152/',
+ 'info_dict': {
+ 'id': 'liga-dos-ultimos-152',
+ 'ext': 'mp4',
+ 'title': 'Liga dos Últimos – RTP Arquivos',
+ 'description': 'Magazine desportivo, com apresentação de Álvaro Costa e comentários em estúdio do professor Hernâni Gonçalves e do sociólogo João Nuno Coelho. Destaque para os jogos de futebol das equipas dos escalões secundários de Portugal, com momentos dos jogos: Agrário de Lamas vs Pampilhoense e Apúlia vs Fragoso.'
},
}, {
'url': 'https://www.rtp.pt/play/p510/aleixo-fm',
@@ -40,8 +81,11 @@ class RTPIE(InfoExtractor):
title = self._html_search_regex(r'<title>(.+?)</title>', webpage, 'title')
+ if 'Este episódio não se encontra disponível' in title:
+ raise ExtractorError('Episode unavailable', expected=True)
+
# Replace irrelevant text in title
- title = title.replace(' - RTP Play - RTP', '')
+ title = re.sub(r' - ?RTP Play - RTP', '', title)
# Check if it's a video split in parts, if so add part number to title
part = self._html_search_regex(r'section\-parts.*<span.*>(.+?)</span>.*</ul>', webpage, 'part', default=None)
@@ -52,69 +96,94 @@ class RTPIE(InfoExtractor):
js_object = self._search_regex(r'(?s)RTPPlayer *\( *({.+?}) *\);', webpage, 'player config')
json_string_for_config = ''
+ filekey_found = False
- # Verify JS object since it isn't pure JSON and maybe it needs some decodings
+ # Verify JS object since it isn't pure JSON and probably needs some fixing
for line in js_object.splitlines():
stripped_line = line.strip()
- # If JS object key is 'file'
- if re.match('file ?:', stripped_line):
+ # If JS object key is 'fileKey'
+ if re.match('fileKey ?:', stripped_line):
+ filekey_found = True
if 'decodeURIComponent' in stripped_line:
- # 1) The file URL is inside object and with HLS encoded...
- hls_encoded = re.match(r"[^[]*\[([^]]*)\]", stripped_line).groups()[0]
- hls_encoded = hls_encoded.replace('"', '').replace('\'', '').replace(',', '')
+ # 1) The value is an encoded URL
+ encoded_url = re.match(r"[^[]*\[([^]]*)\]", stripped_line).groups()[0]
+ encoded_url = re.sub(r'[\s"\',]', '', encoded_url)
+
if 'atob' in stripped_line:
- decoded_file_url = compat_b64decode(
+ # Most of the times 'atob' approach is used but not always so we need to be sure
+ decoded_url = compat_b64decode(
compat_urllib_parse_unquote(
- hls_encoded.replace('"', '').replace(',', ''))).decode('utf-8')
+ encoded_url)).decode('utf-8')
else:
- decoded_file_url = compat_urllib_parse_unquote(hls_encoded)
+ # If no 'atob' we just need to unquote it
+ decoded_url = compat_urllib_parse_unquote(encoded_url)
- # Workaround for new behaviour
- decoded_file_url = decoded_file_url.replace('streaming-vod.rtp.pt/hls/', 'streaming-ondemand.rtp.pt/').replace('.mp4/', '/')
-
- # Insert the decoded HLS file URL into pure JSON string
- json_string_for_config += '\nfile: "' + decoded_file_url + '",'
+ # Insert the (relative) decoded URL in JSON
+ json_string_for_config += f'\nfileKey: "{decoded_url}",'
else:
- # 2) ... or the file URL is not encoded so keep it that way
- json_string_for_config += '\n' + line
+ # 2) ... or the value URL is not encoded so keep it that way
+ json_string_for_config += f'\n{stripped_line}'
- elif not stripped_line.startswith("//") and not re.match('fileKey ?:', stripped_line) and not re.match('.*extraSettings ?:', stripped_line):
- # Ignore commented lines, 'fileKey' entry since it is no longer supported by RTP and also 'extraSettings'
- json_string_for_config += '\n' + line
+ elif (
+ not stripped_line.startswith("//")
+ and not re.match('.*extraSettings ?:', stripped_line)
+ and (not filekey_found or (filekey_found and not re.match('file ?:', stripped_line)))
+ ):
+ # Ignore commented lines and 'extraSettings'. Also ignore 'file' if 'fileKey' already exists
+ json_string_for_config += f'\n{stripped_line}'
# Finally send pure JSON string for JSON parsing
config = self._parse_json(json_string_for_config, video_id, js_to_json)
- # Check if file URL is directly a string or is still inside object
- if isinstance(config['file'], str):
+ if 'fileKey' in config:
+ # 'fileKey' has priority over 'file' on our end
+ file_url = config['fileKey']
+ elif 'file' in config:
+ # 'RTP Arquivos' still uses old regular non-encoded 'file' key
file_url = config['file']
else:
- file_url = config['file']['hls']
+ raise ExtractorError('No valid media source found in page')
ext = determine_ext(file_url)
- if ext == 'm3u8':
- # Download via m3u8 file
+ if ext == 'mp4':
+ # Due to recent changes, we need to hardcode the URL like this and download it using 'm3u8'
+ file_url = f'https://streaming-vod.rtp.pt/hls{file_url}/index-v1-a1.m3u8'
+
+ formats = self._extract_m3u8_formats(
+ file_url, video_id, 'mp4', 'm3u8_native',
+ m3u8_id='hls')
+ elif ext == 'm3u8':
+ # It can be downloaded without any further changes
formats = self._extract_m3u8_formats(
file_url, video_id, 'mp4', 'm3u8_native',
m3u8_id='hls')
-
- self._sort_formats(formats)
else:
+ # Need to set basepath
+ file_url = f'https://cdn-ondemand.rtp.pt{file_url}'
formats = [{
'url': file_url,
'ext': ext,
}]
- if config.get('mediaType') == 'audio':
+ if config['mediaType'] == 'audio':
for f in formats:
f['vcodec'] = 'none'
+ subtitles = {}
+ if 'vtt' in config:
+ sub_lang, sub_lang_full, sub_url = config['vtt'][0]
+ subtitles.setdefault(sub_lang, []).append({
+ 'url': sub_url,
+ 'ext': 'vtt',
+ })
+
return {
'id': video_id,
'title': title,
'formats': formats,
- 'description': self._html_search_meta(['description', 'twitter:description'], webpage),
- 'thumbnail': config.get('poster') or self._og_search_thumbnail(webpage),
+ 'subtitles': subtitles,
+ 'description': self._html_search_meta(['og:description', 'description', 'twitter:description'], webpage),
+ 'thumbnail': config['poster'] or self._og_search_thumbnail(webpage),
}
From 1058c29c74453e80ad918889a0101894a370c9b7 Mon Sep 17 00:00:00 2001
From: vallovic <vallovic@gmail.com>
Date: Sun, 31 Oct 2021 11:16:05 +0000
Subject: [PATCH 9/9] Make extractor compatible with Python 2.6+
---
youtube_dl/extractor/rtp.py | 12 ++++++------
1 file changed, 6 insertions(+), 6 deletions(-)
diff --git a/youtube_dl/extractor/rtp.py b/youtube_dl/extractor/rtp.py
index ffa915075..1c84e9e8d 100644
--- a/youtube_dl/extractor/rtp.py
+++ b/youtube_dl/extractor/rtp.py
@@ -90,7 +90,7 @@ class RTPIE(InfoExtractor):
# Check if it's a video split in parts, if so add part number to title
part = self._html_search_regex(r'section\-parts.*<span.*>(.+?)</span>.*</ul>', webpage, 'part', default=None)
if part:
- title = f'{title} {part}'
+ title = '{title} {part}'.format(title=title, part=part)
# Get JS object
js_object = self._search_regex(r'(?s)RTPPlayer *\( *({.+?}) *\);', webpage, 'player config')
@@ -120,10 +120,10 @@ class RTPIE(InfoExtractor):
decoded_url = compat_urllib_parse_unquote(encoded_url)
# Insert the (relative) decoded URL in JSON
- json_string_for_config += f'\nfileKey: "{decoded_url}",'
+ json_string_for_config += '\nfileKey: "{decoded_url}",'.format(decoded_url=decoded_url)
else:
# 2) ... or the value URL is not encoded so keep it that way
- json_string_for_config += f'\n{stripped_line}'
+ json_string_for_config += '\n{stripped_line}'.format(stripped_line=stripped_line)
elif (
not stripped_line.startswith("//")
@@ -131,7 +131,7 @@ class RTPIE(InfoExtractor):
and (not filekey_found or (filekey_found and not re.match('file ?:', stripped_line)))
):
# Ignore commented lines and 'extraSettings'. Also ignore 'file' if 'fileKey' already exists
- json_string_for_config += f'\n{stripped_line}'
+ json_string_for_config += '\n{stripped_line}'.format(stripped_line=stripped_line)
# Finally send pure JSON string for JSON parsing
config = self._parse_json(json_string_for_config, video_id, js_to_json)
@@ -149,7 +149,7 @@ class RTPIE(InfoExtractor):
if ext == 'mp4':
# Due to recent changes, we need to hardcode the URL like this and download it using 'm3u8'
- file_url = f'https://streaming-vod.rtp.pt/hls{file_url}/index-v1-a1.m3u8'
+ file_url = 'https://streaming-vod.rtp.pt/hls{file_url}/index-v1-a1.m3u8'.format(file_url=file_url)
formats = self._extract_m3u8_formats(
file_url, video_id, 'mp4', 'm3u8_native',
@@ -161,7 +161,7 @@ class RTPIE(InfoExtractor):
m3u8_id='hls')
else:
# Need to set basepath
- file_url = f'https://cdn-ondemand.rtp.pt{file_url}'
+ file_url = 'https://cdn-ondemand.rtp.pt{file_url}'.format(file_url=file_url)
formats = [{
'url': file_url,
'ext': ext,